Introduction/Hypothesis Flow-based therapy via small bore nasal cannula (e.g., high-velocity nasal insufflation (HVNI)) has been shown to be a therapeutic alternative to non-invasive positive pressure ventilation (NIPPV) in undifferentiated respiratory failure. A randomized, multicenter study was performed to evaluate the efficacy of HVNI (Precision Flow; Vapotherm, Inc, Exeter, NH) in comparison to NIPPV, specifically bi-level positive airway pressure, for treatment of patients with COPD presenting with moderate to severe acute hypercapnic respiratory failure (AHRF). The hypothesis tested was HVNI is comparable to NIPPV in providing relief of dyspnea within 4 hours. Methods Adults with known or suspected COPD presenting to 1 of 7 enrolling emergency departments with AHRF defined as an initial PCO2 of 60mmHg or higher and pH of 7.0-7.35 on venous blood gas were screened. Exclusions included need for emergent intubation, severe metabolic derangements, primary condition of congestive heart failure, and presence of significant pneumonia on chest imaging. Subjects were randomized 1:1 to either treatment with HFNI or NIPPV using a full facemask. Assessments occurred at baseline, 30 minutes, 1 hour, and 4 hours after initiation of treatment and included vital signs, subject-rated perceived dyspnea using the modified Borg Score, clinician-performed Patient Stability Index, and venous PCO2, PO2, pH, bicarbonate, and base excess. Results 68 subjects were enrolled with data for analysis at 7 emergency departments and randomized to HVNI (n=36) or NIPPV (n=32). Baseline characteristics were similar. Treatment failure occurred in 4 of 36 subjects (11.1 percent) In the HVNI group and 6 of 32 subjects (18.8percent) in the NIPPV group (p=0.498). Preliminary analysis of the data demonstrated that over 4 hours, mean PCO2 fell from 77.8 ± 13.6mmHg to 68.7 ± 13.9mmHg in the HVNI cohort and 76.5 ± 13.6mmHg to 69.0 ± 20.3mmHg in the NIPPV cohort. The absolute change in PCO2 was similar between the 2 groups (HVNI: -10.6 ± 13.3mmHg vs NIPPV: -8.5 ± 14.1mmHg, p=0.563). Conclusions Treatment of acute hypercapnic respiratory failure due to COPD with HVNI results in similar decreases in venous PCO2 compared to treatment with NIPPV. These results suggest that with similar physiological benefit, treatment with HVNI is a reasonable alternative to NIPPV. This may lead to reduced rates of intubation for patients unable to tolerate NIPPV.